東華大學圖書館 |

Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales // Chuxuan Li.
Author:	Li, Chuxuan,
Description:	1 electronic resource (252 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Contained By:	Dissertations Abstracts International85-11B.
Subject:	Hydrologic sciences. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31242095
ISBN:	9798382762524

Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales /
Li, Chuxuan,

Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales /Chuxuan Li. - 1 electronic resource (252 pages)

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

Landslides constitute widespread and hazardous natural disasters globally, often triggered by heavy rainfall, rapid snowmelt, earthquakes, and anthropogenic activities. Among landslides triggered by various factors, rainfall-triggered landslides have the most pervasive impact worldwide. California in the United States is particularly susceptible to rainfall-triggered landslides due to its variable topography and frequent atmospheric rivers (ARs) during the wet season. In addition, wildfires that are prevalent during the dry season have further increased California's susceptibility to postfire debris flows. Following precipitation, the surface runoff within and downstream of burned areas is substantially higher than unburned land due to reduced vegetation cover and soil infiltration rates. This increased runoff accelerates erosion, leading to an increased risk of triggering debris flows. Furthermore, the diverse climatic zones within California, ranging from mountains with seasonal snowpack, evergreen forests to deserts, coupled with the complex interactions among climatic, meteorological, hydrological, and geological factors, have posed substantial challenges in predicting rainfall-triggered landslides. As climate change is projected to substantially alter the abovementioned causative factors in California, the activities and behaviors of rainfall-triggered landslides are also projected to change. As such, tools that can resolve the various influencing factors and accurately predict rainfall-induced landslide hazards in a rapidly evolving climate are critically needed to tackle the landslide risks.To date, substantial progress has been achieved in predicting rainfall-triggered landslides at individual catchment scales using heuristic, data-driven, and physics-based methods. However, limited attention has been given to advancing prediction models for rainfall-triggered landslides, especially postfire debris flows, at regional scales. Considering the expansive spatial scales of weather systems that trigger landslides and their landfall locations, coupled with the wide spatial extent over which climatic factors operate, the regional scale emerges as the most relevant spatial scale from the perspectives of hazard warning and mitigation as well as climate adaptation. Data-driven approaches are economical and efficient to apply at regional scales, but they encounter two significant challenges due to insufficient representation of physical processes: 1) they only marginally integrate hydrological processes which are considered as the primary drivers for rainfall-triggered landslides and 2) they struggle to generalize to out-of-sample scenarios. In contrast, physics-based, spatially distributed hydrologic models can be applied at regional scales to simulate the relevant hydrological processes contributing to landslides. However, the use of such models to study or predict rainfall-triggered landslides at regional scales remains extremely limited. To address this research gap, this Ph.D. dissertation focuses on augmenting and applying the hydrologic core of the National Oceanic and Atmospheric Administration (NOAA)'s National Water Model, WRF-Hydro, to simulate the hydrological conditions related to the susceptibility of rainfall-triggered landslides including postfire debris flows at regional scales in California. This Ph.D. dissertation consists of five chapters. Chapter One reviews the state of the science and the current challenges in rainfall-triggered landslide prediction techniques. Chapter Two focuses on improving WRF-Hydro's model performance to simulate landslide-relevant hydrological variables (i.e., soil moisture and streamflow). By updating WRF-Hydro's soil parameters using data from two state-of-the-science soil databases, WRF-Hydro's soil moisture and streamflow simulations significantly improve in a central California domain that is characterized by complex topography and frequent landslide occurrence. Chapter Three focuses on augmenting WRF-Hydro to simulate postfire debris flow susceptibility during an AR event in the Dolan burn scar along the Big Sur Coast, California, in January 2021. High susceptibility areas identified using a WRF-Hydro peak discharge based index correspond well with remotely sensed and field observed debris flow occurrences, showing the great potential of WRF-Hydro to predict these hydro-geomorphic hazards at regional scales. Chapter Four investigates the hydro-meteorological conditions underlying massive landslide events triggered by the AR storm series during December 2022-January 2023 across California. According to the water balance status calculated using WRF-Hydro simulated variables and the Multi-Radar/Multi-Sensor (MRMS) Precipitation forcing, most of (91%) the observed landslide occurrences are associated with positive water balance status. A K-means Clustering on four hydro-meteorological variables reveals that the landslides are dominated by three distinct hydrological processes. Finally, Chapter Five provides a conclusion and outlines future directions to advance the predictive capabilities for rainfall-triggered landslides at regional scales. This includes a discussion on the applicability of the methodological framework developed in this Ph.D. research to forecast and project landslide hazards in the future. In summary, the Ph.D. dissertation presents pioneering studies that augment and utilize a physics-based hydrological model to study and hindcast the susceptibility of rainfall-triggered landslides including postfire debris flows at regional scales. It establishes a methodological framework that can be extended to address various needs including operational weather forecasts and climate projections. Through ongoing development and integration with state-of-the-science remote sensing and machine learning techniques, the developed framework will stand as a compelling physics-based component to advance the broad field of climate-driven land surface hazards.

English

ISBN: 9798382762524Subjects--Topical Terms:

3168407
Hydrologic sciences.
Subjects--Index Terms:

Atmospheric rivers

Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales /
LDR:07552nmm a22004573i 4500 001 2435592
005 20251009122839.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798382762524
035 $a (MiAaPQD)AAI31242095
035 $a AAI31242095
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Li, Chuxuan, $e author. $0 (orcid)0000-0001-7759-0377 $3 3825938
245 1 0 $a Augmentation and Use of WRF-Hydro to Predict Rainfall-Triggered Landslide Susceptibility at Regional Scales / $c Chuxuan Li.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (252 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
500 $a Advisors: Horton, Daniel Committee members: Buscarnera, Giuseppe; Beddows, Patricia; Wang, Jiali; Handwerger, Alexander.
502 $b Ph.D. $c Northwestern University $d 2024.
520 $a Landslides constitute widespread and hazardous natural disasters globally, often triggered by heavy rainfall, rapid snowmelt, earthquakes, and anthropogenic activities. Among landslides triggered by various factors, rainfall-triggered landslides have the most pervasive impact worldwide. California in the United States is particularly susceptible to rainfall-triggered landslides due to its variable topography and frequent atmospheric rivers (ARs) during the wet season. In addition, wildfires that are prevalent during the dry season have further increased California's susceptibility to postfire debris flows. Following precipitation, the surface runoff within and downstream of burned areas is substantially higher than unburned land due to reduced vegetation cover and soil infiltration rates. This increased runoff accelerates erosion, leading to an increased risk of triggering debris flows. Furthermore, the diverse climatic zones within California, ranging from mountains with seasonal snowpack, evergreen forests to deserts, coupled with the complex interactions among climatic, meteorological, hydrological, and geological factors, have posed substantial challenges in predicting rainfall-triggered landslides. As climate change is projected to substantially alter the abovementioned causative factors in California, the activities and behaviors of rainfall-triggered landslides are also projected to change. As such, tools that can resolve the various influencing factors and accurately predict rainfall-induced landslide hazards in a rapidly evolving climate are critically needed to tackle the landslide risks.To date, substantial progress has been achieved in predicting rainfall-triggered landslides at individual catchment scales using heuristic, data-driven, and physics-based methods. However, limited attention has been given to advancing prediction models for rainfall-triggered landslides, especially postfire debris flows, at regional scales. Considering the expansive spatial scales of weather systems that trigger landslides and their landfall locations, coupled with the wide spatial extent over which climatic factors operate, the regional scale emerges as the most relevant spatial scale from the perspectives of hazard warning and mitigation as well as climate adaptation. Data-driven approaches are economical and efficient to apply at regional scales, but they encounter two significant challenges due to insufficient representation of physical processes: 1) they only marginally integrate hydrological processes which are considered as the primary drivers for rainfall-triggered landslides and 2) they struggle to generalize to out-of-sample scenarios. In contrast, physics-based, spatially distributed hydrologic models can be applied at regional scales to simulate the relevant hydrological processes contributing to landslides. However, the use of such models to study or predict rainfall-triggered landslides at regional scales remains extremely limited. To address this research gap, this Ph.D. dissertation focuses on augmenting and applying the hydrologic core of the National Oceanic and Atmospheric Administration (NOAA)'s National Water Model, WRF-Hydro, to simulate the hydrological conditions related to the susceptibility of rainfall-triggered landslides including postfire debris flows at regional scales in California. This Ph.D. dissertation consists of five chapters. Chapter One reviews the state of the science and the current challenges in rainfall-triggered landslide prediction techniques. Chapter Two focuses on improving WRF-Hydro's model performance to simulate landslide-relevant hydrological variables (i.e., soil moisture and streamflow). By updating WRF-Hydro's soil parameters using data from two state-of-the-science soil databases, WRF-Hydro's soil moisture and streamflow simulations significantly improve in a central California domain that is characterized by complex topography and frequent landslide occurrence. Chapter Three focuses on augmenting WRF-Hydro to simulate postfire debris flow susceptibility during an AR event in the Dolan burn scar along the Big Sur Coast, California, in January 2021. High susceptibility areas identified using a WRF-Hydro peak discharge based index correspond well with remotely sensed and field observed debris flow occurrences, showing the great potential of WRF-Hydro to predict these hydro-geomorphic hazards at regional scales. Chapter Four investigates the hydro-meteorological conditions underlying massive landslide events triggered by the AR storm series during December 2022-January 2023 across California. According to the water balance status calculated using WRF-Hydro simulated variables and the Multi-Radar/Multi-Sensor (MRMS) Precipitation forcing, most of (91%) the observed landslide occurrences are associated with positive water balance status. A K-means Clustering on four hydro-meteorological variables reveals that the landslides are dominated by three distinct hydrological processes. Finally, Chapter Five provides a conclusion and outlines future directions to advance the predictive capabilities for rainfall-triggered landslides at regional scales. This includes a discussion on the applicability of the methodological framework developed in this Ph.D. research to forecast and project landslide hazards in the future. In summary, the Ph.D. dissertation presents pioneering studies that augment and utilize a physics-based hydrological model to study and hindcast the susceptibility of rainfall-triggered landslides including postfire debris flows at regional scales. It establishes a methodological framework that can be extended to address various needs including operational weather forecasts and climate projections. Through ongoing development and integration with state-of-the-science remote sensing and machine learning techniques, the developed framework will stand as a compelling physics-based component to advance the broad field of climate-driven land surface hazards.
546 $a English
590 $a School code: 0163
650 4 $a Hydrologic sciences. $3 3168407
650 4 $a Geography. $3 524010
650 4 $a Geomorphology. $3 542703
650 4 $a Atmospheric sciences. $3 3168354
653 $a Atmospheric rivers
653 $a California
653 $a Debris flows
653 $a Hydrological modeling
653 $a Landslides
653 $a Wildfires
690 $a 0388
690 $a 0366
690 $a 0484
690 $a 0725
710 2 $a Northwestern University. $b Earth and Planetary Sciences. $e degree granting institution. $3 3825939
720 1 $a Horton, Daniel $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-11B.
790 $a 0163
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31242095